Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <CA+G9fYuifLivwhCh33kedtpU=6zUpTQ_uSkESyzdRKYp8WbTFQ@mail.gmail.com>
 <ZJLzsWsIPD57pDgc@FVFF77S0Q05N> <ZJQXdFxoBNUdutYx@FVFF77S0Q05N>
 <CA+G9fYtAutjL3KpZsQyJuk4WqS=Ydi2iyVb5jdecZ-SOuzKCmA@mail.gmail.com> <CANk7y0h+oXNhUzTFQ_Wy-iySRdBi0ezu1Y_hOBtAxmK5AG4dgA@mail.gmail.com>
In-Reply-To: <CANk7y0h+oXNhUzTFQ_Wy-iySRdBi0ezu1Y_hOBtAxmK5AG4dgA@mail.gmail.com>
From:   Puranjay Mohan <puranjay12@gmail.com>
Date:   Mon, 3 Jul 2023 11:36:53 +0200
Message-ID: <CANk7y0gi4o+4U6c9QmnDCJDRXNM_98or_4tO-dHOEwZ4fj3gkw@mail.gmail.com>
Subject: Re: next: Rpi4: Unexpected kernel BRK exception at EL1
To:     Naresh Kamboju <naresh.kamboju@linaro.org>
Cc:     Mark Rutland <mark.rutland@arm.com>,
        Linux ARM <linux-arm-kernel@lists.infradead.org>,
        open list <linux-kernel@vger.kernel.org>,
        linux-rpi-kernel@lists.infradead.org,
        Netdev <netdev@vger.kernel.org>, lkft-triage@lists.linaro.org,
        Arnd Bergmann <arnd@arndb.de>,
        Dan Carpenter <dan.carpenter@linaro.org>,
        Linus Walleij <linus.walleij@linaro.org>,
        Catalin Marinas <catalin.marinas@arm.com>,
        Will Deacon <will@kernel.org>,
        Anshuman Khandual <anshuman.khandual@arm.com>,
        Song Liu <song@kernel.org>, Alexei Starovoitov <ast@kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Hi Naresh,

On Mon, Jun 26, 2023 at 11:04=E2=80=AFAM Puranjay Mohan <puranjay12@gmail.c=
om> wrote:
>
> Hi Naresh,
>
> On Thu, Jun 22, 2023 at 2:35=E2=80=AFPM Naresh Kamboju
> <naresh.kamboju@linaro.org> wrote:
> >
> > Hi Mark,
> >
> > On Thu, 22 Jun 2023 at 15:12, Mark Rutland <mark.rutland@arm.com> wrote=
:
> > >
> > > On Wed, Jun 21, 2023 at 01:57:21PM +0100, Mark Rutland wrote:
> > > > On Wed, Jun 21, 2023 at 06:06:51PM +0530, Naresh Kamboju wrote:
> > > > > Following boot warnings and crashes noticed on arm64 Rpi4 device =
running
> > > > > Linux next-20230621 kernel.
> > > > >
> > > > > Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
> > > > >
> > > > > boot log:
> > > > >
> > > > > [   22.331748] Kernel text patching generated an invalid instruct=
ion
> > > > > at 0xffff8000835d6580!
> > > > > [   22.340579] Unexpected kernel BRK exception at EL1
> > > > > [   22.346141] Internal error: BRK handler: 00000000f2000100 [#1]=
 PREEMPT SMP
> > > >
> > > > This indicates execution of AARCH64_BREAK_FAULT.
> > > >
> > > > That could be from dodgy arguments to aarch64_insn_gen_*(), or else=
where, and
> > > > given this is in the networking code I suspect this'll be related t=
o BPF.
> > > >
> > > > Looking at next-20230621 I see commit:
> > > >
> > > >   49703aa2adfaff28 ("bpf, arm64: use bpf_jit_binary_pack_alloc")
> > > >
> > > > ... which changed the way BPF allocates memory, and has code that p=
ads memory
> > > > with a bunch of AARCH64_BREAK_FAULT, so it looks like that *might* =
be related.
> > >
> > > For the benefit of those just looknig at this thread, there has been =
some
> > > discussion in the original thread for this commit. Summary and links =
below.
> > >
> > > We identified a potential issue with missing cache maintenance:
> > >
> > >   https://lore.kernel.org/linux-arm-kernel/ZJMXqTffB22LSOkd@FVFF77S0Q=
05N/
> > >
> > > Puranjay verified that was causing the problem seen here:
> > >
> > >   https://lore.kernel.org/linux-arm-kernel/CANk7y0h5ucxmMz4K8sGx7qogF=
yx6PRxYxmFtwTRO7=3D0Y=3DB4ugw@mail.gmail.com/
> > >
> > > Alexei has dropped this commit for now:
> > >
> > >   https://lore.kernel.org/linux-arm-kernel/CAADnVQJqDOMABEx8JuU6r_Deh=
yf=3DSkDfRNChx1oNfqPoo7pSrw@mail.gmail.com/
> >
> > Thanks for the detailed information.
> > I am happy to test any proposed fix patches.
>
> I have sent the v4 of the patch series:
> https://lore.kernel.org/bpf/20230626085811.3192402-1-puranjay12@gmail.com=
/T/#t
> This works on my raspberry pi 4 setup. If possible can you test this
> on the similar setup where it was failing earlier?

I think my previous email was missed.
Can you test the V4 series in the same setup?
This is still not applied to the bpf-next tree.

Thanks,
Puranjay.